chore(deps): update ai sdk - #155
Conversation
dawsontoth
left a comment
dawsontoth
left a comment
There was a problem hiding this comment.
Approved — verified locally on c53ee859
Lockfile-only AI-SDK group refresh: @ai-sdk/provider 4.0.7 → 4.0.8, @ai-sdk/provider-utils 5.0.28 → 5.0.32, @ai-sdk/gateway 4.0.59 → 4.0.67.
The group rules are doing their job. npm ls shows one flat copy of everything and no cross-major split — the failure class that install and CI cannot catch:
@ai-sdk/anthropic@4.0.44 / google@4.0.56 / openai@4.0.50 → @ai-sdk/provider@4.0.8 (deduped)
@openai/agents@0.15.0 ─ agents-core@0.15.0 ─ agents-openai@0.15.0 ─ agents-realtime@0.15.0
@openai/agents-extensions@0.15.0 → @ai-sdk/provider@4.0.8, ai@7.0.83 (both deduped)
ollama-ai-provider-v2@4.0.1 → @ai-sdk/provider-utils@5.0.32 (deduped)
Single @openai/agents-core@0.15.0, so lifecycle/getModel.ts is not sitting on a version seam this time.
Runtime probe — aisdk()-wrapped model handed to Agent/run() with an invalid key, which is the path getModel.test.ts mocks out and therefore never exercises:
| provider | spec | result |
|---|---|---|
| anthropic | v4 |
APICallError: API key is invalid. ✅ |
v4 |
APICallError: API key not valid. ✅ |
|
| openai | v4 |
APICallError: Incorrect API key provided ✅ |
Lockfile health (this repo's recurring hazard): keys 1648, react-native* entries 28, harper still 5.2.4 — unchanged from base, npm ci exit 0. This PR does not trip the pruning problem; sibling #156 does, see my review there.
Suite: npm run build exit 0 · npm run lint exit 0 · npm run format (dprint check) exit 0 · npm test 53 files / 345 tests passed.
All 6 CI checks green.
|
🎉 This PR is included in version 0.16.47 🎉 The release is available on GitHub release Your semantic-release bot 📦🚀 |
1 participant
This PR contains the following updates:
4.0.40→4.0.444.0.49→4.0.564.0.45→4.0.507.0.73→7.0.83Release Notes
vercel/ai (@ai-sdk/anthropic)
v4.0.44Compare Source
Patch Changes
3e125ba]v4.0.42Compare Source
Patch Changes
591d25b: feat: add batch completion webhooks.experimental_startTextBatchaccepts awebhookUrl, and the gateway provider registers it through the batchcallbackUrlcontract and exports typed async-job metadata. Direct Anthropic and OpenAI batch providers return an unsupported warning when the option is provided.591d25b]v4.0.41Compare Source
Patch Changes
b74971f]vercel/ai (@ai-sdk/google)
v4.0.56Compare Source
Patch Changes
3ad9da9: Preserve complete Google Generative Language usage metadata in raw usage results.e9bc618: Omit unsupported frequency and presence penalties from Gemini 2.5 requests and return warnings instead.v4.0.55Compare Source
Patch Changes
56d492f: Surface prompt-level Google safety blocks without candidates as content-filter results with prompt feedback metadata.v4.0.54Compare Source
Patch Changes
1f7835c: feat (provider/google, provider/google-vertex): Gemini 3.5 Transcribe support — unary transcription (gemini-3.5-transcribe) via generateContent with language detection, speaker diarization, word timestamps, and custom vocabulary, plus streaming transcription (gemini-3.5-transcribe-live) over the Live API WebSocket withmode: 'VERBATIM' | 'SMART'transcription formattingv4.0.53Compare Source
Patch Changes
3e125ba]v4.0.51Compare Source
Patch Changes
e7fc90e: feat(google): support the Gemini Batch API with experimental_startTextBatch591d25b]v4.0.50Compare Source
Patch Changes
b74971f]vercel/ai (@ai-sdk/openai)
v4.0.50Compare Source
Patch Changes
e6a2992: Fix Responses history serialization for regular functions namedtool_search.v4.0.49Compare Source
Patch Changes
2523403: Normalize non-object replayed Chat Completions tool arguments to empty objects.3e125ba]v4.0.47Compare Source
Patch Changes
591d25b: feat: add batch completion webhooks.experimental_startTextBatchaccepts awebhookUrl, and the gateway provider registers it through the batchcallbackUrlcontract and exports typed async-job metadata. Direct Anthropic and OpenAI batch providers return an unsupported warning when the option is provided.2214258: Prevent negative text output token counts when providers report reasoning tokens. Perplexity reasoning tokens are now treated as separate from completion tokens.591d25b]v4.0.46Compare Source
Patch Changes
b74971f]vercel/ai (ai)
v7.0.83Compare Source
Patch Changes
8dd86a9: Validate persisted typed tool calls against current input and output schemas.Schema-incompatible empty or error inputs and terminal history from unavailable
tools remain loadable as dynamic tool parts instead of exposing unvalidated
values under current static tool types.
fda13b3: Allow chats to continue automatically after tool approval denials reach theoutput-deniedstate.957146c: add operation-level outcomes to UI message stream end callbacksce6849a: fix(ai): handle stitchable stream cancellation before an inner stream is registeredv7.0.82Compare Source
Patch Changes
3e125ba: Allow manual tool approval statuses to include a reason and preserve it acrosscore, model, and UI approval requests. OPA
requires-approvaldecisions nowsurface their reason to human approvers. UI request chunks serialize the
optional
reason, while UI messages retain it asapproval.requestReasonseparately from an approver's response
reason.0e7994c]3e125ba]v7.0.79Compare Source
Patch Changes
b251584: Preserve active text and reasoning parts when another merged UI message stream finishes a step, and align workflow stream normalization with the explicit part end chunks.591d25b: feat: add batch completion webhooks.experimental_startTextBatchaccepts awebhookUrl, and the gateway provider registers it through the batchcallbackUrlcontract and exports typed async-job metadata. Direct Anthropic and OpenAI batch providers return an unsupported warning when the option is provided.9de0baf: Parse structured generateText output when providers omit finishReason but return text.591d25b]v7.0.78Compare Source
Patch Changes
96970bb: Continue approvedgenerateText,streamText, andWorkflowAgentturns with a model-visible tool error when revalidated tool input is invalid.e12e068]v7.0.77Compare Source
Patch Changes
b74971f]a371615]v7.0.76Compare Source
Patch Changes
c6d57f3: fix(ai): prevent duplicate text and reasoning part ids677a707: fix(ai): allow nullish metadata schemas for UI messages with branded IDsv7.0.74Compare Source
Patch Changes
5941bd6]9a4337d]Configuration
📅 Schedule: (in timezone America/New_York)
🚦 Automerge: Enabled.
♻ Rebasing: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.
👻 Immortal: This PR will be recreated if closed unmerged. Get config help if that's undesired.
This PR was generated by Mend Renovate. View the repository job log.